Exendins are a family of peptides that lower blood glucose levels and have some sequence similarity (53%) to GLP-1 (Goke et al., 1993 J. Biol. Chem. 268; 19650-19655; incorporated herein by reference). The exendins are found in the venom of the Gila Monster (Heloderma suspectum; Raufman 1996 Reg. Peptides, 61; 1-18; incorporated herein by reference) and the Mexican beaded lizard. Exendin-4 found in the venom of the Gila Monster is of particular interest. The Exendin-4 precursor is a 87 amino acid polypeptide that includes signal and prosequences (Sequence ID No: 1). It is processed to give the amidated 39 amino acid Exendin-4 (Sequence ID No: 2). Published PCT international patent application, WO 98/35033, which is incorporated herein by reference, discloses the cDNA encoding proexendin peptide, including exendin and other novel peptides, its isolation, and antibodies which specifically recognize such peptides (Pohl et al. 1998, J. Biol. Chem. 273; 9778-9784; incorporated herein by reference). Exendin-4 has been shown to be a strong agonist of the GLP-1 receptor in isolated rat insulinoma cells. It also has a greater biological half-life compared to GLP-1 (Young et al 1999 Diabetes 48; 1026-1034; incorporated herein by reference). This may be expected since the His-Ala domain of active GLP-1 (Sequence ID No: 3) recognized by DPP-IV is not present in Exendin-4, which has the sequence His-Gly instead.
Exendin-4 given systemically lowers blood glucose levels by 40% in diabetic db/db mice (WO 99/07404; incorporated herein by reference). U.S. Pat. No. 5,424,286 by Eng, which is incorporated herein by reference, discloses that a considerable portion of the N-terminal sequence is essential in order to preserve insulinotropic activity.
The number of amino acid residues in a peptide sequence has a dramatic influence on the production costs of peptides made by solid phase synthetic chemistry. The cost of manufacturing a 50-mer peptide is at least 5 times greater than the cost of a 10-mer peptide. Solid phase peptide synthesis also requires the use of corrosive solvents such as TFS and hydrofluoric acids, a number of synthetic steps are required to produce the peptide, and subsequent purification of the peptide. Therefore, there is a need for an efficient and cost effective method for producing large quantities of biologically active Exendin-4 and other insulinotropic peptides.
Expression of a recombinant protein or polypeptide in Pichia or other microorganisms does not always result in successful production of full length polypeptide. Often, the heterologous recombinant polypeptide/protein are subjected to cleavage/degradation by proteases intracellularly. Given the structure of a protein and the multiplicity of proteases present in the cell, specificity of the amino acid sequence recognized by or targeted by the protease is in many cases undetermined and not published. It is therefore not always possible for a person skilled in the art to guess which protease would be responsible for cleavage/degradation of a specific recombinant polypeptide/protein. For eg., it has been reported that expression of murine or human endostatin in Pichia led to a product which was missing C-terminal lysine (Folkman et al, 1999 May; 15(7):563-72). When the Pichia homologue of KEX-1 gene of Saccharomyces cerevisiae was disrupted, the Pichia host secreted full length endostatin into the medium. In another study it was reported that disruption of KEX-2, but not YPS-1 gene, in Pichia allowed production of mammalian gelatin, which was being cleaved at monoargylinic sites of the protein (Werten and de Wolf, Applied and Environmental Microbiology, 2005 May; 71(5):2310-7).
The instant invention proves the prevention of in vivo proteolytic cleavage of proteins/polypeptide having the amino acids HG (His-Gly) at the N-terminus with the disruption of Saccharomyces cerevisiae STE13 gene homolog of Pichia. Very specifically the problem associated with proteolytic cleavage of Glycine-extended Exendin-4 (GlyExendin-4, Exendin-4 precursor with C-terminal glycine) has been shown to be solved by disruption of Saccharomyces cerevisiae STE13 gene homolog of Pichia.
GLYEXENDIN-4 (Sequence ID No: 4)HGEGTFTSDLSKQMEEEAVRLFIEWLKNGGPSSGAPPPSG